Model comparison

harrier-oss-v1-270m vs ced-base

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

harrier-oss-v1-270mced-base
Parameters268M86M
Architecturegemma3ced
Layers18
Native context32,768
Mixture of expertsnono
Quantizations published103
Smallest quantization0.23 GiB0.12 GiB
Q4_K_M0.24 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
Contextharrier-oss-v1-270mced-baseRatio
4,0960.07 GiB
8,1920.14 GiB
16,3840.28 GiB
32,7680.56 GiB
65,5361.13 GiB
131,0722.25 GiB